FY2011 has seen significant progress toward accomplishing all of the Specific Aims. For Aim 1, we have demonstrated that magnetic susceptibility-weighted imaging at ultra-high field strength (7 T) is sensitive to the presence and orientation of myelinated white matter tracts, a prerequisite for demonstrating that this technique can detect demyelination (12, 17). Specifically, our work has shown that myelinated fiber bundles contain a component of rapidly decaying magnetization that is quantifiable, and that the decay time constant can vary by as much as 50% when the tissue is rotated with respect to the MRI system. We believe on theoretical and experimental grounds that these findings are due to the structure and contents of myelin sheets in white matter. Preliminary results from experiments performed under this Aim also indicate that the diffusion properties of intra-axonal metabolites are deranged in multiple sclerosis and suggest that the directionality of metabolite diffusion may be a more straightforward readout of axonal damage than the directionality of water diffusion (manuscript in preparation). This raises the possibility that damaged but not yet moribund axons can be detected and quantified, which would potentially allow assessment of the efficacy of putative neuroprotective drugs in clinical trials. For Aim 2, we have uncovered previously unknown dynamics of blood-brain-barrier opening in newly developing MS lesions (3, 14), providing new insights into the processes by which those lesions form and opening new avenues for investigating the mechanisms of action of drugs that modulate blood-brain-barrier permeability. Specifically, we have shown that the blood-brain barrier is initially open in the center of newly forming lesions, which probably corresponds to the small, inflamed vein around which these lesions form. Under these conditions, serum contents fill the lesion from the center to the periphery, a pattern we call centrifugal dynamics. Subsequently, as the lesions expand, the location of blood-brain-barrier opening moves to the lesions periphery, which histologically corresponds to the most active area of inflammation. Under these conditions, serum contents fill the lesion from the periphery to the center (centripetal dynamics). We interpret these differences as evidence of distinct pathological processes, perhaps corresponding to tissue damage and repair, and further work in patients and animal models is investigating this hypothesis. These experiments underline the importance of neurovascular interactions in the pathogenesis if multiple sclerosis. For Aim 3, we have developed new and fully automated image segmentation techniques that can identify the volumes of healthy and diseased portions of the brain, including gray matter, white matter, and lesions (1, 11, 13). In a cross-sectional analysis of a multiple sclerosis cohort, this analysis revealed an inverse correlation between white matter volume in the brain and clinical disability scores, despite the absence of frank white matter atrophy (manuscript under review). We suspect that this finding may be related to more successful tissue repair, perhaps even remyelination, in some individuals. We have also demonstrated that MRI techniques, including diffusion-weighted and magnetization transfer-weighted imaging, are sensitive to MS-related tissue damage, and that they can be used in a clinically relevant way to quantify progressive tissue damage over time (2, 9, 10, 15, 16). Specific applications in 2011 have been to the visual system, a portion of the brain that is especially susceptible to multiple sclerosis-related tissue damage and that can now be interrogated by a variety of imaging and clinical techniques. Finally, we have developed powerful statistical tools that take advantage of the rich spatiotemporal information available in longitudinal MRI studies in order more accurately to assess that damage (4-8). In particular, a longitudinal analysis of yearly scans in 78 people with multiple sclerosis revealed that these techniques could be used as outcome measures in Phase II clinical trials with relatively small sample sizes (on the order of 40 people per arm) (9). This is one of the first confirmations that such advanced techniques can feasibly be used for clinical trial purposes.